Geometric Decoupling: Diagnosing the Structural Instability of Latent

Yuanbang Liang; Zhengwen Chen; Yu-Kun Lai

cs.CVApr 20, 2026

Geometric Decoupling: Diagnosing the Structural Instability of Latent

Authors: Yuanbang Liang, Zhengwen Chen, Yu-Kun Lai

Organizations: School of Computer Science and Informatics, Cardiff University, Cardiff, United Kingdom

Abstract

Latent Diffusion Models (LDMs) achieve high-fidelity synthesis but suffer from latent space brittleness, causing discontinuous semantic jumps during editing. We introduce a Riemannian framework to diagnose this instability by analyzing the generative Jacobian, decomposing geometry into \textit{Local Scaling} (capacity) and \textit{Local Complexity} (curvature). Our study uncovers a \textbf{Geometric Decoupling"}: while curvature in normal generation functionally encodes image detail, OOD generation exhibits a functional decoupling where extreme curvature is wasted on unstable semantic boundaries rather than perceptible details. This geometric misallocation identifies Geometric Hotspots" as the structural root of instability, providing a robust intrinsic metric for diagnosing generative reliability.

Explore similar work

Intrinsic-Hybrid Latent Diffusion Models for Generative Modeling on Unknown Manifolds
Aug 5, 2026Yizhu Wang, Mu Niu, Xiaochen YangIdentity-Conditioned Latent Diffusion Models Latent Diffusion Model
How to Train Your Latent Diffusion Language Model Jointly With the Latent Space
May 8, 2026Viacheslav Meshchaninov, Alexander Shabalin, Egor Chimbulatov +4Diffusion Language Models Latent Diffusion Model
One Pass Is Not Enough: Recursive Latent Refinement for Generative Models
May 14, 2026Mehdi Esmaeilzadeh, Alexia Jolicoeur-Martineau, Chirag Vashist +1Large-Scale Image Generation Generative Models

Aug 5, 2026 · stat.MLJ/K move · Enter open · S save

Intrinsic-Hybrid Latent Diffusion Models for Generative Modeling on Unknown Manifolds

Yizhu Wang, Mu Niu, Xiaochen Yang

We introduce the Intrinsic Hybrid Latent Diffusion Model (ILDM), a generative framework that integrates probabilistic dimensionality reduction with geometry-aware diffusion on unknown manifolds. While diffusion models (DMs) have achieved state-of-the-art results in high-dimensional data synthesis, they rely on large training datasets and ignore intrinsic geometric structure. Latent diffusion models (LDMs) address the high dimensionality by learning a latent space, but they typically impose a Euclidean structure, failing to capture the underlying manifold geometry, especially problematic in data-sparse regimes. ILDM addresses these limitations by interpreting the latent space as a chart of an unknown Riemannian manifold, with geometry and uncertainty quantified through a probabilistic decoder. The forward process is a hybrid diffusion that switches between Riemannian and Euclidean dynamics based on local uncertainty, where the Riemannian component is governed by a probabilistic metric tensor derived from the decoder. To learn the generative dynamics, we introduce an approximate denoising score matching method tailored to the hybrid diffusion setting, enabling a backward process defined by hybrid Langevin dynamics. Experiments on COIL-100, MNIST, and cardiac MRI datasets demonstrate that ILDM significantly improves generation quality, achieving lower FID and LPIPS scores compared to standard diffusion and latent diffusion models.

Identity-Conditioned Latent Diffusion Models Latent Diffusion Model Generative Models Manifolds Dimensionality Reduction

Open paper Ask about it PDF

Abstract

Explore similar work

Intrinsic-Hybrid Latent Diffusion Models for Generative Modeling on Unknown Manifolds

How to Train Your Latent Diffusion Language Model Jointly With the Latent Space

One Pass Is Not Enough: Recursive Latent Refinement for Generative Models